Local Weighted Semi-supervised Discriminant Analysis for Dimensionality Reduction
Honghua Wang, Yumei Sun, Hongxiu Li, Mao Jun Zhou · 2015
In this paper, we present a novel weighted version of semi-supervised discriminant analysis method by assigning weights to each labeled samples. The proposed within-class weight can detect the outliers and between-class weight can discover the support points in boundaries between different classes. In addition, our proposed method is robust to diverse-density classes and imbalanced boundaries. For high-dimensional dataset, our method can find a nice low-dimensional projection to preserve the discriminative information and manifold structure embedded in both labeled and unlabeled samples. It can also be easily kernelized to form a nonlinear method and do semi-supervised induction. The experiments show that our method can achieve very promising classification accuracies than other methods.